Tubelab SE: Removing MOSFETs? - Page 3 - diyAudio
Go Back   Home > Forums > Commercial Sector > Vendor Forums > Tubelab
Home Forums Rules Articles diyAudio Store Gallery Wiki Blogs Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Tubelab Discussion and support of Tubelab products, prototypes and experiments

Reply
 
Thread Tools Search this Thread
Old 12th July 2016, 04:12 PM   #21
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
Not much new here - really hot this time of year out in the radio room.

Played around last week with some Rad Shack 70 volt line transformers, afaik, they are not air gapped, but at these low currents I'm not sure that makes much of a difference.

Performance was not that bad, low frequency response was markedly better than the OPT I have on it now, so I may swap them out. SMPTE IMD was about the same. I got best results with 46 and this transformer at what I calculated to be the 4K tap. I used the next tap down to energize the second grid. My real interest here was trying these as an OPT in a choke loaded configuration, but I don't seem to have any suitable chokes on hand. I suspect that at these low currents, a good line transformer is probably quite OK on its own.

I also jumpered in one of the 5K 10 watt Transcendars. No UL tap, so triode only - needless to say every aspect of performance was markedly better with this high quality transformer.

Win W5JAG

edit: I would also add, this thing remains hum free, which was totally unexpected to me. The 6.3 volt heater is not referenced to B+, or to ground with a virtual center tap. I've been meaning to do the latter. There is no provision for hum reduction on the 2.5 volt heater.

I jumpered in the Transcendar, just to see if it would go low enough to hum. It reproduces well below 60 Hz with that transformer, but there is no hum in my test speaker. It is capable of reproducing hum.

Last edited by w5jag; 12th July 2016 at 04:29 PM.
  Reply With Quote
Old 27th July 2016, 03:03 PM   #22
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
Line transformers have caught my attention lately, for some reason.

In the first picture is an Atlas Sound HT-87. These are advertised as being made from audio grade laminations ( whatever that means ) and have a claimed frequency response of 50-15,000 Hz + 1 dB, and an 8 watt output rating. They weigh a bit more than a pound; the second picture shows it next to a typical ( Rad Shack ) 70 volt transformer. These are somewhat expensive for line transformers, but I got a case of them off eBay for $24 shipped to my door, so I figured it was worth a gamble to try them out. I am hoping that I can find a balanced center point on one of the taps so they can be used as push pull OPT's.

In the meantime, I rigged one up to the mule. I measured their frequency response, at 1 volt RMS input to the mule, as flat down to 40 Hz. It may go lower, I didn't try to measure below that. ( I also need to re measure the OPT that is currently on the mule, I think I've been doing it wrong and erroneously stated its low frequency response ).

For power levels up to 150 mw output or so, the HT-87 performance was as good as or better than the air gapped OPT that is on the mule. The two tone IMD went up pretty rapidly at higher outputs, however. This was using the 1 watt tap on the plate, and 2 watt tap on the second grid - all other combinations of taps were unacceptable. I also used it as a triode on the 1 watt tap with acceptable results.

I also tried choke loading the mule with the HT-87, even though I really don't have any suitable chokes. I set it up as a triode, and started sticking chokes in series with the plate. I used a .47uF output capacitor - values below that didn't show much output. Results were dismal below about 10 Henries. Results got better with increasing inductance up to about 23.5 Henries, the limit of chokes I had at the house to string together. Finally, I just used the primary of a 5K 10 watt Transcendar, which I think is 30 Henries, getting rid of all the mess and about six feet of jumpers in the process. I really didn't see any two tone IMD improvement by doing this. I think between about 20 and 30 henries is probably adequate for something like this.

At power levels to 100 mw output, choke loading did not provide any two tone IMD improvement. Above that level, it did show significant improvement, effectively doubling the power output, with no increase in the two tone IMD. If one could get suitable chokes and good line transformers on the cheap, this could be a viable methed of making an inexpensive low power amp. paying full price for chokes and line OPT's would probably be as much as just getting a Hammond 125xSE type transformer in the first place.

Win W5JAG

edit: what I call choke loading, the audiophiles call para feed. if that clears up any confusion.
Attached Images
File Type: jpg IMG_20160726_195012.jpg (473.0 KB, 283 views)
File Type: jpg IMG_20160726_195000.jpg (448.3 KB, 278 views)
File Type: jpg IMG_20160726_194921.jpg (548.0 KB, 263 views)

Last edited by w5jag; 27th July 2016 at 03:08 PM.
  Reply With Quote
Old 29th July 2016, 08:14 PM   #23
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
Default The Atlas Sound HT-87 as a push pull OPT

Quote:
... I am hoping that I can find a balanced center point on one of the taps so they can be used as push pull OPT's.
They work well as a low power push pull OPT.

I have an Antique Electronic Supply 11BM8 push pull amp that has, for the last decade, resided under an eave on the deck at our primary residence. It's seen some hard times, including multiple hose blasts while in operation.

I pulled it, thinking I might liberate some good OPT's, freeing them up for something else, and replace them with mediocre OPTS more suitable for an outdoor deck amp. They are identical size with the stock OPT's.

The results were unexpected. My notes are in the last picture. The left channel was horrific with the stock OPT's, causing me to wonder if sheltered outdoor exposure and the occasional hose blast have damaged it. It looks OK. The right channel was somewhat better, but nothing to brag about. I did not record those measurements.

The HT-87 was a vast improvement on the left channel. It was a significant improvement on the right channel, but not as dramatic, probably due to tube mis match. The tubes in the left channel are the last of my "matched" 11BM8's. The tubes in the right channel are just random pulls from my warehouse to replace water blasted tubes.

The frequency response measurement was done on the right channel, at 1 watt output, since I expected it would be the worst case measurement. The sine wave is 1 kHz, 6 watts at onset of clipping. The square wave is at 10kHz, 1 watt.

These look to be pretty good as push pull OPT's.

edit: I haven't seen a voltage rating on these - they look to be good at the voltages in this amp and the mule, but they have very little time in operation, and caution is warranted here.

Win W5JAG
Attached Images
File Type: jpg IMG_20160728_183315.jpg (477.5 KB, 254 views)
File Type: jpg IMG_20160728_194809.jpg (852.2 KB, 245 views)
File Type: jpg IMG_20160728_194923.jpg (886.6 KB, 67 views)
File Type: jpg IMG_20160729_074531.jpg (228.9 KB, 81 views)
File Type: jpg IMG_20160728_213427.jpg (271.4 KB, 75 views)

Last edited by w5jag; 29th July 2016 at 08:33 PM.
  Reply With Quote
Old 23rd August 2016, 03:01 PM   #24
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
The line transformer diversion is, temporarily, grinding to a close.

Re: the AES 11BM8 amp, I took off the HT-87's, and replaced them with the Radio Shack #32-1031 10 watt line transformers. The shared cathode resistors were also running scorching hot and had changed value, even though the amp really doesn't have a lot of hours on it, so I replaced those with 5 watt 270 ohm resistors. These still run hotter than I like.

I also verified that the distrortion and frequency response anomalies follow the unmatched tubes. The present B+ voltage is 215 volts, and Pd is 12.5 watts. Both line trannies seem to handle this voltage without difficulty.

The Radio Shack transformer is not as efficient as the Atlas Transformer, nor does it have the low distortion characteristic. Two tone IMD on the matched side was 1.5% at one quarter watt; 2.5% at one half watt, and 3.0% at one watt. Not good at all for a push pull amplifier. Freq resonse was spectacular for a cheap OPT, however. On the matched side, at 1 watt output, it was - 3dB at 35 Hz; -1 dB at 100 Hz; 0 db at 1 khz; 0 dB at 10 kHz; and -1.0 dB at 20 kHz. The bad side was similar, the - 3dB point being 60 Hz.

The square wave looks rough, however, these sound fine. Since they are on the deck amplifier, I've just left them in place.

Not sure if all line transformers work this way, but I've found with both the Atlas and the Rad Shack, the "center tap" is two taps down from the tap in current use. For those interested, the load is calculated thus: the line voltage is squared ( 70.7*70.7 = 5000) and this product is divided by the wattage tap to get a load resistance thus: 5000 / 1 watt = 5000 ohms. The center tap would be two taps down, at 4 watts.

Win W5JAG
Attached Images
File Type: jpg IMG_20160801_ratshack_line_11BM8.jpg (261.0 KB, 52 views)
File Type: jpg IMG_20160801_ratshack_line_sq_11BM8.jpg (905.5 KB, 34 views)
  Reply With Quote
Old 23rd August 2016, 03:21 PM   #25
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
Default If you have a transformer, try it

Since it is easy to do, I've been hooking up anything that has windings to try out.

Filament transformers: work okay, over a narrow frequency range, which, happily covers speech pretty well. I have vague memories as a kid of using 6.3 volt filament transformers to "match' a 600 ohm line output to a typical speaker. Nowadays, I would use a PA line transformer for that.

Small, hamfest type, vintage OPT's: work well, provided relatively poor low frequency response is acceptable. Most are - 3 to -4 dB when you get much below 80 Hz. High frequency response is not a problem.

Push Pull transformers as Single Ended OPT's: Works, surprisingly, at least at these flea powers and currents, probably not at higher levels. Trying to use the end to end plate leads won't work well at all, but using the center tap as the B+lead, and a plate lead as the load, seems to work well, with good dstortion characteristics and good freq response. Trying this with a line transformer won't work at all.

This is the set up I have been using: I pull the 6X5 rectifier, and jumper B+ into the mule from the regulated power supply. I have a 25 watt 1k Clarostat potentiometer as the cathode resistor, and at a given B+, just twist the Clarostat for lowest distortion or highest output, or whatever.

For the 46, with the mule's onboard power supply, best distortion results are at a cathode resistance of 750 ohms. Distortion improves slightly up to about 300 volts B+, beyond which doesn't seem to make much difference.

Win W5JAG
Attached Images
File Type: jpg IMG_20160820_095841.jpg (601.1 KB, 55 views)
  Reply With Quote
Old 19th September 2016, 06:25 PM   #26
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
Default 46 ( and 47 ) in the SSE

This SSE board was last seen running 6CD6's before being removed from its chassis to make way for the 801TSE board.

First pic is the SSE board with the octal sockets cut out, and five pin sockets installed, held in place by wavy washers.

It's easier to experiment with all components on the side of the board with the tubes, thus all but two resistors were so relocated, the resistors were all checked and new caps installed. The second pic shows the board set up with the components for 46 installed in one channel.

Third pic is testing the front end to make sure both sides still work properly before proceeding further - the pic shows each half of the 12AT7 putting about 75 volts RMS into a 1 meg load, 1 kHz input, about 300 volts to the CCS.

Fourth pic is the mule power supply jumpered to the SSE board, a Transcendar 5K OPT for the load.

Contrary to my intital belief, directly heated tubes on the SSE board looks to have real merit. I may follow through with a complete amp build. The TSE has nothing to fear from this, but it sure looks like it will be possible to out perform, by a large margin, a low power SSE built with indirectly heated tubes.

This was doing 0.8 % IMD at 1/10 watt, increasing to about 5% at 4/10 watt, with a 12AU7 / 46 combination. There is room for improvement yet to come. To put that in perspective, a well respected transformer winder, using new low IMD transformers and low IMD circuitry, got 2% at 1 watt from a Type 50 tube, IIRC, and my TSE 801 got down to 4% ( I can get it lower running the tubes harder ) at 1 watt with an RF final tube and cheap Hammond universal OPT's.

Win W5JAG
Attached Images
File Type: jpg IMG_20160917_150932.jpg (553.1 KB, 61 views)
File Type: jpg IMG_20160919_092817.jpg (674.4 KB, 65 views)
File Type: jpg IMG_20160918_182457.jpg (604.2 KB, 64 views)
File Type: jpg IMG_20160918_194313.jpg (702.7 KB, 62 views)
  Reply With Quote
Old 24th September 2016, 11:37 PM   #27
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
Moving on, here is 307A / VT-225 in the mule. The perched transformer I picked up at the Joplin Hamfest; it should drop right in and do 400 volts, more or less, under load.

I have a half dozen used 307A, all came from Fair Radio for not much money. The first one I tried lit up like a neon sign inside the plate structure but, after a few hours, that resolved itself and it seems to work normally. 307A #2 has some wierd kind of on / off pulsation - mostly off - doesn't seem like oscillation, but I'm not ruling that out yet. 307A #3 is pictured, and it has a filament issue - one leg is visibly brighter as can be seen in the photo. It's burning in now to see if that will clear. I haven't tried the other three yet.

This is just a quickie circuit to get it going - I haven't measured anything except filament voltage. A 0.51 ohm resistor in each leg set it at 5.6 volts AC. Cathode bias. G2 and G3 are tied to plate. 470K grid resistor.

Win W5JAG
Attached Images
File Type: jpg IMG_20160924_175702.jpg (468.7 KB, 55 views)

Last edited by w5jag; 24th September 2016 at 11:40 PM.
  Reply With Quote
Old 25th September 2016, 09:11 PM   #28
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
Well, I don't have any 300B's, but it is already very apparent 307A is one absolutely terrific sounding tube.

It's a real shame they are so scarce.

Win W5JAG
  Reply With Quote
Old 26th September 2016, 06:59 PM   #29
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
Here is the underhood shot.

As a starter, I had the cathode resistor and bypass hooked to one leg of the heater. This worked, but hummed unacceptably at 8 mW with no signal input. I made a virtual center tap with two 22 ohm resistors, and connected the resistor and bypass to that, and this lowered the hum to 0.02 mW at idle. A lot better, but still barely audible with my ear against the test speaker. A real balance pot may be needed with more sensitive speakers.

The OPT wiring is wackadoo. Carelessly, I connected the screen/supressor to the 7K tap, and the anode to the 5K tap. It was some time before I noticed this obvious mistake. It seems to work really well this way. Maybe this is the way to get negative feedback to the screens with this OPT, and I was doing it wrong before.

Even with the crumby OPT, the mule sounds terrific wired this way, with this tube. As a utility amp, I'm not inclined to change anything about it. And to think, I was just going to stick an 807 in the socket. I may still, but I'm glad I tried these first.

Win W5JAG
Attached Images
File Type: jpg IMG_20160925_164918.jpg (470.2 KB, 51 views)
File Type: jpg IMG_20160925_184837.jpg (464.2 KB, 46 views)
  Reply With Quote
Old 11th October 2016, 05:13 PM   #30
w5jag is offline w5jag  United States
diyAudio Member
 
Join Date: Feb 2007
Location: Arkansas
I was going to replace the PT on the mule with the larger one, but, even though the "chassis" is properly cut, this seemed like a fair amount of work, and I'm lazy, and, for a simple utility amp, not really needed.

The blindingly obvious solution, which for years I've managed to not notice ( like that Yaesu amp on my bench that has apparently not moved in years ) was to use one of my regulated high voltage bench supplies.

So I rigged up a cable with an eight pin plug to plug into the mule rectifier socket, emulating the pin out of the 6X5. The 6X5 / 6AX5 uses five of the eight octal pins. I used one of the free pins to tie the chassis/grounds together. I used another pin to bring in the adjustable negative bias for fixed bias. Banana plugs are expensive, and I'm cheap as well as lazy, so I used the test leads from old, free, HF VOM's that I had long ago blown up. The bias lead is longer with a shrink marker for positive identification.

The connection at the plug looks ( and is ) unacceptably fragile. It needs to be potted for strength, but I'm not sure what to use here. Suggestions welcomed; voltage is 400 max.

This will also work on SSE / TSE PCB's, provided they are modified for the 6AX5 pin out. Two jumpers are required; they are identified in the child resistant SSE thread. Or another specific umbilical could be made. This is needed as I am probably going to make a DHT SSE amp, possibly using 307A, maybe 46 or 47. Still undecided; but the board is done up and ready for direct experimenting, freeing up the mule for other exploits.

With a few new holes, the mule is now sporting an octal socket. I also have some loctal and 12 pin compactron sockets that will fit without more drilling, and with the 2.5 volt transformer gone, a new one can be installed for offbeat filament voltages.

Win W5JAG
Attached Images
File Type: jpg IMG_20161006_145458.jpg (478.0 KB, 32 views)
File Type: jpg IMG_20161006_145539.jpg (362.9 KB, 28 views)
File Type: jpg IMG_20161006_153350.jpg (540.0 KB, 34 views)
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Tubelab SE zxx222 Tubelab 172 4th September 2016 12:53 PM
Mosfets for Tubelab SE? mctavish Tubelab 54 9th January 2015 12:18 AM
F/S Tubelab SE stonehome Swap Meet 14 25th December 2014 10:56 PM
Tubelab SE vs Simple SE Pez Tubelab 25 29th October 2011 02:16 AM
Need Help with Tubelab SE lth1 Tubelab 9 1st October 2010 09:06 PM


New To Site? Need Help?

All times are GMT. The time now is 06:25 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Copyright 1999-2017 diyAudio

Content Relevant URLs by vBSEO 3.3.2
Wiki